New pathway for heterogenization of molecular catalysts by non-covalent interactions with carbon nanoreactors

Article abstract of DOI:10.1021/cm502986d

A novel approach to heterogenization of catalytic molecules is demonstrated using the nanoscale graphitic step edges inside hollow graphitized carbon nanofibers (GNFs). The presence of the fullerene C₆₀ moiety within a fullerene-salen Cu^II complex is essential for anchoring the catalyst within the GNF nanoreactor as demonstrated by comparison to the analogous catalyst complex without the fullerene group. The presence of the catalyst at the step edges of the GNFs is confirmed by high-resolution transmission electron microscopy (TEM) and energy-dispersive X-ray spectroscopy (EDX) with ultraviolet/visible (UV/vis) spectroscopy, demonstrating only negligible (ca. 3%) desorption of the fullerene-salen Cu^II complex from the GNFs into solution under typical reaction conditions. The catalyst immobilized in GNFs shows good catalytic activity and selectivity toward styrene epoxidation, comparable to the analogous catalyst in solution. Moreover, the fullerene-salen Cu^II complex in GNFs demonstrates excellent stability and recyclability because it can be readily separated from the reaction mixture and employed in multiple reaction cycles with minimal loss of activity, which is highly advantageous compared to catalysts not stabilized by the graphitic step edges that desorb rapidly from GNFs.

Full text of DOI:10.1021/cm502986d

Article
pubs.acs.org/cm
New Pathway for Heterogenization of Molecular Catalysts by Non-
covalent Interactions with Carbon Nanoreactors
Maria A. Lebedeva,† Thomas W. Chamberlain,† Martin Schroder,† and Andrei N. Khlobystov*,†,‡
̈
‡
†School of Chemistry, and Nottingham Nanotechnology and Nanoscience Centre, University of Nottingham, University Park,
Nottingham NG7 2RD, United Kingdom
S
* Supporting Information
ABSTRACT: A novel approach to heterogenization of catalytic molecules is demonstrated using the nanoscale graphitic step
edges inside hollow graphitized carbon nanofibers (GNFs). The presence of the fullerene C60 moiety within a fullerene−salen
CuII complex is essential for anchoring the catalyst within the GNF nanoreactor as demonstrated by comparison to the analogous
catalyst complex without the fullerene group. The presence of the catalyst at the step edges of the GNFs is confirmed by high-
resolution transmission electron microscopy (TEM) and energy-dispersive X-ray spectroscopy (EDX) with ultraviolet/visible
(UV/vis) spectroscopy, demonstrating only negligible (ca. 3%) desorption of the fullerene−salen CuII complex from the GNFs
into solution under typical reaction conditions. The catalyst immobilized in GNFs shows good catalytic activity and selectivity
toward styrene epoxidation, comparable to the analogous catalyst in solution. Moreover, the fullerene−salen CuII complex in
GNFs demonstrates excellent stability and recyclability because it can be readily separated from the reaction mixture and
employed in multiple reaction cycles with minimal loss of activity, which is highly advantageous compared to catalysts not
stabilized by the graphitic step edges that desorb rapidly from GNFs.
including immobilization of catalysts on organic and inorganic
supports, such as polystyrenes and other polymers,4 silica,5
mesoporous materials,6 and zeolites,7 by trapping the catalyst
molecules inside porous hosts,8 and via formation of hydrogen
bonds, metal coordination,9 or electrostatic interactions10
between the catalyst molecule and the support material.
Non-covalent heterogenization of molecular catalysts also
presents several important challenges. Non-covalent interac-
tions are usually reversible, resulting in the gradual loss of the
catalytic activity upon recycling.11 Hence, a good anchoring
group is required to provide sufficient binding of the catalyst to
the surface of the support to afford high stability of the resulting
heterogeneous material. Additionally, non-covalent interactions
often lower the degree of control over the location of the
catalyst molecules on the support surface. Carbon nanostruc-
tures, such as carbon nanotubes (CNTs) and graphite, have
been used extensively as robust supporting materials for
INTRODUCTION
■
The ability to separate and recycle the metal catalyst from
reaction mixtures without excessive purification steps is crucial
for sustainable transition-metal-based catalysis.1 Recyclability of
catalysts can be enhanced by heterogenization of the catalyst
molecules, achieved by immobilization on solid supports. This
improves the stability of the molecular catalyst and, moreover,
renders it insoluble in commonly used organic solvents, thus
offering a straightforward separation of the catalyst from the
reaction mixture.2 Heterogenization of a molecular catalyst can
be achieved by immobilization on a variety of supports using
covalent or non-covalent binding between the catalyst
molecules and the support material. The preparation of
covalently linked catalysts can be synthetically very demanding,
requiring the preparation of non-symmetric ligands bearing
functional groups that can form chemical bonds with the
support material.3 On the other hand, non-covalent immobi-
lization can, in principle, be achieved more readily and results in
the formation of materials in which the structure and the
intrinsic properties of the molecular catalyst are retained. Non-
covalent catalyst binding has been achieved in a variety of ways,
Received: August 14, 2014
Revised: October 20, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/cm502986d | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
or GNF, 5 mg) was dissolved or suspended in MeCN (1 mL)
containing styrene (52 mg, 0.5 mmol), and a solution of BuOOH in
immobilization of catalysts on both internal and external
surfaces and provide excellent stability and thermal and electric
conductivity. For example, extended conjugated systems of sp2
carbons of CNTs and graphite allow for catalyst molecules
tagged with polyaromatic hydrocarbons, such as pyrene, to be
anchored to the surface using π−π stacking interactions.12
However, such anchoring is relatively weak, low-directional, and
not site-specific, with the catalyst deposited randomly over the
carbon surface.
In this study, we report a new methodology for
heterogenization of transition metal catalysts by tagging them
with a fullerene group, which allows for reliable and site-specific
anchoring inside graphitized carbon nanofibers (GNFs) via
strong van der Waals interactions between the fullerene cage
and the internal graphitic step edges of GNFs. The spherical
shape and large π-electron system of functionalized C60
fullerenes is known to have a high affinity for sp2-hybridized
carbon structures and is an excellent match for the preferential
binding at graphitic step edges rather than the flat terrace of the
inner surface of the GNF support.13 We demonstrate herein
that non-covalent immobilization of a fullerene-tagged [Cu-
(salen)] catalyst inside GNF nanoreactors can be used to form
well-defined heterogeneous catalysts with molecules located in
a very distinct, confined environment. These heterogeneous
catalysts exhibit enhanced stability and recyclability, while
retaining the activity and selectivity of the individual catalytic
centers.
t
nonane (5.5 M, 0.27 mL, 1.5 mmol) was added. The reaction mixture
was heated to 80 °C for 7 h, and aliquots of the reaction mixture were
then analyzed by 1H NMR spectroscopy in CDCl3 solution after each
hour without further purification. All quoted reaction yields are a result
of at least three repeat reactions.
1
Styrene, a: H NMR (300 MHz, CDCl3) δ: 7.30−7.50 (m, 5H, Ar
H), 6.69 (dd, J = 17.6 and 10.9 Hz, 1H, CHCH2), 5.72 (d, J = 17.6
Hz, 1H, CHCH2), 5.22 (d, J = 10.9 Hz, 1H, CHCH2).
Styrene epoxide, b: 1H NMR (300 MHz, CDCl3) δ: 7.30−7.50 (m,
5H, Ar H), 3.86 (m, 1H, −CH(O)−), 3.15 (dd, J = 5.5 and 4.1 Hz,
1H, CH(O)−CH2), 2.80 (dd, J = 5.5 and 2.6 Hz, 1H, CH(O)−CH2).
1
c: H NMR (300 MHz, CDCl3) δ: 7.20−7.40 (m, 5H, Ar H), 5.28
(dd, J = 8.0 and 3.6 Hz, 1H), 4.08−4.24 (m, 2H).
d: 1H NMR (300 MHz, CDCl3) δ: 7.20−7.40 (m, 5H, Ar H), 5.03
(d, J = 7.7 Hz, 1H), 4.08−4.24 (m, 2H).
1
e: H NMR (300 MHz, CDCl3) δ: 10.02 (s, 1H, CHO), 7.20−
7.40 (m, 5H, Ar H), 5.09 (s, 2H).
Leaching Test under Catalytic Conditions. The catalyst
(1@GNF, 2.5 mg, 0.08 μmol or 2@GNF, 2.5 mg, 0.08 μmol) was
suspended in MeCN (0.5 mL) containing styrene (26 mg, 0.25
t
mmol), and in some cases, a solution of BuOOH in nonane (5.5 M,
0.13 mL, 1.5 mmol) was added. The reaction mixture was heated to 80
°C for 7 h and filtered hot to remove the insoluble materials. The
filtrate was then diluted to 5 mL and analyzed using UV/vis
spectroscopy.
Recyclability Test. Styrene epoxidation was performed as
described above. After the reaction mixture was heated to 80 °C for
7 h, the reaction mixture was allowed to cool to room temperature and
the catalyst was removed by filtration, washed extensively with cold
CH3CN (20 mL), and dried under vacuum for 20 h. The resulting
material was then used in the next catalytic cycle.
EXPERIMENTAL SECTION
■
Compounds 1 and 2 were synthesized according to the previously
reported procedures.15 GNFs (PR19 Pyrograf, chemical vapor
deposition) were thermally annealed at 400 °C for 30 min prior to
use. Styrene was passed through neutral alumina shortly before all
reactions. All other reagents and solvents were purchased from Sigma-
Aldrich and used without further purification. Reactions were carried
out under an aerobic atmosphere and monitored by 1H nuclear
magnetic resonance (NMR) spectroscopy. 1H NMR spectra were
recorded using a Bruker DPX 300 spectrometer. Ultraviolet/visible
(UV/vis) spectra were measured using a Lambda 25 PerkinElmer
spectrometer.
RESULTS AND DISCUSSION
■
Transition metal complexes of salen derivatives are established
as highly successful and versatile catalysts. The scope of
reactions catalyzed by metal−salen complexes is extremely wide
and includes carbon−carbon bond formation, heteroatom−
heteroatom bond formation, and carbon−heteroatom bond
formation reactions.14 We have synthesized the fullerene-tagged
CuII salen complex 1 (Figure 1) in six steps according to our
Preparation of Heterogeneous Catalysts 1@GNF and
2@GNF. GNFs (5 mg) were annealed at 400 °C for 30 min and
immersed into a solution of the corresponding [Cu(salen)] complex
(1, 0.2 mg, 0.16 μmol or 2, 0.1 mg, 0.16 μmol) in tetrahydrofuran
(THF) (2 mL). The mixture was treated by ultrasound and stirred for
30 min, and the solvent was slowly removed under reduced pressure. A
new portion of THF (2 mL) was added, and the procedure was
repeated 4 times to ensure complete encapsulation of the molecules in
GNFs. Upon completion, the 1@GNF and 2@GNF catalysts were
dried under vacuum for 20 h.
High-Resolution Transmission Electron Microscopy
(HRTEM) Characterization. HRTEM imaging was performed using
a JEOL 2100F transmission electron microscope (field emission
electron gun, information limit of 0.19 nm) using an accelerating
voltage of 200 kV. Transmission electron microscopy (TEM)
specimens were prepared by casting several drops of a methanolic
suspension of 1@GNF or 2@GNF onto a nickel TEM specimen grid
mounted “lacey” carbon film and dried under a stream of nitrogen.
Energy-dispersive X-ray (EDX) spectra were recorded for isolated
nanofibers of 1@GNF or 2@GNF using an Oxford Instruments X-ray
detector at 200 kV.
Leaching Test. 1@GNF (5 mg, 0.16 μmol of Cu complex) or
2@GNF (5 mg, 0.16 μmol of Cu complex) was immersed in MeCN
(4 mL) and heated to 50 °C in a sealed UV/vis cell, and absorption
spectra of the resulting solution were measured every 10 min.
Styrene Epoxidation. The catalyst (1, 0.2 mg, 0.16 μmol; 2, 0.1
mg, 0.16 μmol; 1@GNF, 5 mg, 0.16 μmol; 2@GNF, 5 mg, 0.16 μmol;
Figure 1. Structure of the fullerene-derivatized [Cu(salen)] complex
(1) and its fullerene-free analogue (2).
recently reported procedure.15 The analogous fullerene-free
complex 2 was prepared for comparison. [CuII(salen)]
complexes are known to catalyze epoxidation reactions of
alkenes, and it has been demonstrated previously that a slight
enhancement of the catalytic activity of the transition metal
complex is observed upon addition of the fullerene cage
because of the electron-withdrawing nature of the carbon
cage.15
B
dx.doi.org/10.1021/cm502986d | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 2. (a) Schematic representation of 1@GNF showing fullerene-containing molecules located preferentially on the step edges of the GNF
(fullerene molecules and GNF are not drawn to scale). TEM images of 1@GNF show (b) the accessible internal channel of the nanoreactor, (c) the
internal surface of a GNF filled with 1 (white circle), and (d) a high-magnification TEM image of 1@GNF showing an individual molecule of 1
located on the step edge of the GNF (top) and its schematic representation (bottom). (e) EDX spectroscopy confirms the presence of copper within
the 1@GNF structure (the Ni peaks are due to the TEM specimen grid).
Immobilization and encapsulation of fullerene-containing
molecules inside single-, double-, and multi-walled CNTs has
been extensively studied, and the procedure to achieve high
filling yields is now well-established.16 In contrast, GNFs have
not been used widely for fullerene encapsulation because of
their very wide internal diameters (50−70 nm). However, the
structure of GNF consists of stacked truncated cones of
graphite layers, and this creates a sequence of step edges
(Figure 2a), which are ideally suited to accommodate catalyst
molecules. Indeed, it has been demonstrated previously for
various metal nanoparticles that guest species are located
predominantly on these step-edge sites.17−19 Furthermore, the
diameters of GNFs are sufficiently large to minimize diffusion
resistance of the reactant and product molecules to and from
the catalytic centers.
The heterogeneous catalyst, 1@GNF, was prepared by
introducing the fullerene-tagged molecules into the internal
channel of GNF using a solution method. Initially, the GNFs
were thermally treated at 400 °C to remove any residual water
molecules from their internal channels and then immersed into
a saturated solution of 1 in THF, followed by slow removal of
the solvent under reduced pressure to deposit the molecules
onto the GNF. Fresh solvent was then added to dissolve any
molecules on the outside surface of the GNF, and this
procedure was repeated several times to ensure a high degree of
pore filling. Immobilization of control compound 2 into GNF
was carried out using the same method. The structures of the
resultant composite materials (1@GNF and 2@GNF; Figure
2a) were analyzed by HRTEM (panels b and c of Figure 2).
HRTEM images of 1@GNF show the presence of the
individual fullerene-containing molecules (Figure 2c, observed
as circles with a diameter of ca. 0.7 nm) located on the internal
step edges of the GNF, so that the surface of contact between
the C60 moiety and the step edge of GNF is maximized (Figure
2d). The presence of the [Cu(salen)] moiety is more difficult
to visualize because the majority of organic molecules rapidly
decompose under the electron beam of TEM. However, the
presence of copper inside the GNF was confirmed by localized
EDX spectroscopy (Figure 2e). The structure of 2@GNF was
also analyzed by HRTEM, and although fullerene-free
molecules were not resolved in the micrograph, the presence
of the copper-containing species was also confirmed by EDX.
This procedure enables deposition of the molecules
predominantly on the inside of the GNF and also enables
precise control of the total amount of catalyst loaded into the
GNF support. The catalyst loading in both 1@GNF and
2@GNF was 0.03 mmol of copper salen catalyst/g of GNF,
which is comparable to similar immobilized metal catalysts in
CNTs attached using either covalent20−22 or non-covalent
bonding.23,24 Table S1 of the Supporting Information compares
the catalyst loading in the present study with systems reported
previously. The efficiency of the catalyst immobilization in
GNFs was quantified by comparing the amount of catalyst that
leached from the GNFs into solution over time (Figure 3). The
same amount of freshly prepared 1@GNF or 2@GNF with
identical catalyst loading was immersed in an excess of MeCN
and heated to 50 °C in a sealed vessel. The resulting solution
was analyzed by UV/vis spectroscopy over time (Figure 3).
The majority of the fullerene-containing molecules (97%) was
retained in the GNF, giving an almost colorless solution (left
C
dx.doi.org/10.1021/cm502986d | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
and radical addition products c and d. The formation of
epoxide is promoted by CuII centers via formation of a CuIII
peroxo metallocycle intermediate, which then breaks to yield
the styrene oxide and regenerate the CuII complex.25
Indeed, in the absence of catalyst, only 2% of styrene epoxide
is formed (compound b in Scheme 1 and Table 1) and the
major products are compounds c (39%) and d (12%) (Scheme
1), formed as a result of an undesirable radical addition of tert-
butyl peroxide radicals to the double bond of styrene.26 In the
presence of empty GNFs, a similar trend is observed in which
the epoxide b is a minor product (8% yield) and formation of
compounds c and d is favored (50 and 22%, respectively).
Introducing catalysts 1 or 2 promotes epoxide formation,
preferentially forming product b (42−46%) rather than c (15−
26%) and d (29−20%). This activity and selectivity are
comparable to the majority of catalytic systems reported before
for styrene epoxidation (see Table S2 of the Supporting
Information). The formation of e, which is a product of
rearrangement of epoxide b, depends upon the amount of
epoxide in the reaction mixture. However, it is always a minor
product, and the total yield of e only reaches a maximum of
∼10%. The yield and rate of the reaction catalyzed by fullerene-
containing complex 1 is slightly higher than the reaction
catalyzed by fullerene-free complex 2 (green and orange curves,
respectively, in Figure 4). This is the result of the electron-
Figure 3. Plot showing the amount of the metal complex leached into
MeCN at 50 °C over time for 1@GNF (purple line) and 2@GNF
(blue line). The inset shows photographs of the resultant mixtures of
1@GNF (left) and 2@GNF (right) in MeCN after 5 h.
inset in Figure 3). In comparison, the fullerene-free complex 2
rapidly leached from the GNF support to form a bright yellow
solution (right inset in Figure 3), with only 30% of the initial
complex retained in the GNF after 5 h. This confirms that
attachment of the fullerene cage to the active [Cu(salen)]
catalyst has a dramatic effect on host−guest interactions within
the pores of the support and, consequently, the stability of the
resulting composite materials.
The performance of the heterogeneous catalysts 1@GNF
and 2@GNF was compared in a model styrene epoxidation
reaction along with the molecules 1 and 2 under homogeneous
t
Scheme 1. Epoxidation of Styrene (a) by BuOOH in the
Presence of [Cu(salen)] Forming Styrene Epoxide (b), 1-
Phenyl-1,2-bis-tert-butylperoxy Ether (c), 1-Phenyl-1-tert-
butylperoxy-2-tert-butyloxy Ether (d), and 2-
Phenylacetaldehyde (e)
Figure 4. Rate of formation of styrene epoxide from styrene in the
presence of 1 (green), 2 (orange), 1@GNF (purple), 2@GNF (blue),
and GNF (gray) and in the absence of catalyst (red).
t
Table 1. Catalytic Epoxidation of Styrene with BuOOH in
MeCN at 80 °C Showing the Distribution of Compounds a−
withdrawing effect of the fullerene cage depleting the electron
density on the CuII center and promoting its catalytic activity,
consistent with previous observations.15
e after 7 h
yield (%)
Complete conversion of styrene a and a 42% yield of epoxide
b is achieved in the presence of the heterogeneous catalyst
1@GNF in 7 h, an outcome very similar to the homogeneous
reaction. The heterogeneously catalyzed reaction, however,
occurs noticeably slower than the reaction in solution (purple
and green curves, respectively, in Figure 4). This is attributed to
the fact that the catalytic centers are less accessible in 1@GNF
because of the confinement inside the GNF nanoreactor. As a
result, complete conversion of styrene and formation of similar
amounts of styrene epoxide are achieved over a slightly longer
period compared to the solution reaction (Figure 4). This
difference in the reaction rates of 1 and 1@GNF also confirms
that the catalyst molecules remain immobilized on the GNF
throughout the whole process under these conditions. In
comparison, under the same reaction conditions, 2@GNF
catalyst
a
b
c
d
e
1
2
0
0
0
0
46
42
42
43
8
2
15
26
19
14
50
39
2
2
2
1
2
2
29
20
29
31
22
12
2
2
2
2
2
1
10
12
10
12
4
2
2
1
1@GNF
2@GNF
GNF
2
1
1
1
16
46
2
3
1
0.5
0.5
no catalyst
2
0.5
1
conditions (Scheme 1 and Table 1). Styrene was reacted with
tBuOOH in dry MeCN at 80 °C under an aerobic atmosphere,
and the reaction mixture was monitored over time by 1H NMR
spectroscopy. The reaction gives a distribution of four different
products (Scheme 1), with the target epoxide b undergoing
further transformations to form the corresponding aldehyde e
D
dx.doi.org/10.1021/cm502986d | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
shows a nearly identical rate of formation of styrene epoxide to
2 in solution (orange and blue curves, respectively, in Figure 4),
indicating that both reactions occur in solution under
homogeneous conditions. Similar trends are observed for the
kinetic curves for styrene conversion and for selectivity for
styrene epoxide monitored over time (see Figures S2 and S3 of
the Supporting Information).
These trends are further supported by the UV/vis spectros-
copy measurements of the reaction mixtures under catalytic
conditions. Leaching tests were carried out in the absence of
tBuOOH to simplify quantitative measurements (see the
Supporting Information for details) but in otherwise identical
conditions to the catalytic reaction, and these confirm that the
majority of the catalyst is leached out of 2@GNF into solution
as a result of only weak interactions between 2 and the GNF. In
contrast, 97% of molecules of 1 are retained on GNF (see
Figure S5 and Table S3 of the Supporting Information).
To highlight the excellent stability of this heterogeneous
catalytic system formed using non-covalent interactions
between the fullerene-tagged 1 and GNF, we studied the
recyclability of 1@GNF over a number of repeat catalytic
cycles. We tested the performance of 1@GNF and 2@GNF in
five consecutive epoxidation reactions of styrene (Figure 5),
occur in solution because of both epoxidation/polymerization
of the fullerene cage and dimerization of the [Cu(salen)]
complex, leading to the formation of inactive dimeric species
and aggregates27 under an aerobic atmosphere at elevated
temperatures.
CONCLUSION
■
A non-covalent supramolecular approach for the heterogeniza-
tion of molecular catalysts has been developed using the unique
affinity of C60 fullerene for carbon nanostructures. A fullerene-
tagged [Cu(salen)] complex has been incorporated into GNFs
to form robust and stable catalytically active nanostructures.
Fullerene−GNF interactions are sufficiently strong to allow for
the catalyst molecules to be retained in the GNFs when heated
in MeCN over many hours. As a result, this material possesses
excellent recyclability while showing a significantly better
activity compared to a catalyst without fullerene over multiple
reaction cycles.
Synthetic methodologies to tag a variety of transition metal
complexes with a fullerene cage,28 including various bipyr-
idine,29 terpyridine,30 and porhyrin31 complexes, have been
extensively developed in the past decade, enabling our
methodology to be expanded to a variety of other existing
catalytic systems. The enhanced catalyst recyclability achieved
in GNFs can be particularly valuable for processes involving
rare metals, such as Ru, Os, Pd, Pt, Ag, and Au, and has the
potential to significantly lower the cost of their usage in
industrial processes.
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional 1H NMR spectra, UV/vis absorption spectra,
styrene epoxidation kinetic curves, and additional literature
data. This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: andrei.khlobystov@nottingham.ac.uk.
■
Figure 5. Comparison of the stability and recyclability of the catalyst
in 1@GNF (purple) and 2@GNF (blue) in five consecutive
epoxidation reaction cycles of styrene for 7 h.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported by the European Research Council
(ERC) (Consolidator Grant to Andrei N. Khlobystov and
Advanced Grant to Martin Schroder), the Engineering and
Physical Sciences Research Council (EPSRC), and the
University of Nottingham.
with each cycle lasting 7 h. The catalyst was separated from the
reaction mixture after each cycle by filtration, extensively
washed with MeCN to remove any traces of the reactants/
products, and used in the next otherwise identical reaction
cycle. The conversion of styrene was 100% for the first catalytic
cycle in each case. An identical 42% yield of epoxide was
achieved for both catalysts in the first cycles of catalysis.
However, the catalytic activity of 2@GNF shows a significant
drop in the second cycle, forming only 27% of styrene epoxide
(blue columns in Figure 5), whereas the fullerene-containing
1@GNF affords a 37% yield of styrene epoxide (purple
columns in Figure 5). A further decrease in activity is observed
for 2@GNF, showing only 22% yield of epoxide after 5 cycles,
whereas in the presence of 1@GNF, even after 5 cycles, the
yield of epoxide remains significantly higher and stabilizes at
34% yield. These results confirm the excellent stability of 1@
GNF, in which the catalytic centers are retained on the support
material for a longer time. In addition, spatial separation of the
fullerene-tagged metal complex on the GNF step edges may
also prevent decomposition of the catalyst molecules, which can
■
̈
REFERENCES
(1) Corma, A.; Garcia, H. Chem. Rev. 2002, 102, 3837.
■
(2) Baleizao, C.; Garcia, H. Chem. Rev. 2006, 106, 3987.
̃
(3) (a) Kim, J. H.; Kim, G. J. Catal. Lett. 2004, 92, 123. (b) Park, D.
W.; Choi, S. D.; Choi, S. J.; Lee, C. Y.; Kim, G. J. Catal. Lett. 2002, 78,
145.
(4) Lu, J.; Toy, P. Chem. Rev. 2009, 109, 815.
(5) (a) Choudary, B.; Ramani, T.; Maheswaran, H.; Prashant, L.;
Ranganath, K.; Kumar, K. Adv. Synth. Catal. 2006, 348, 493.
(b) Kantam, M.; Ramani, T.; Chakrapani, L.; Choudary, B. J. Mol.
Catal. A.: Chem. 2007, 274, 11.
(6) Serrano, D.; Aguado, J.; Vargas, C. Appl. Catal., A 2008, 335, 172.
(7) Guo, X.-F.; Kim, Y.-S.; Kim, G.-J. Top. Catal. 2009, 52, 153.
(8) Corma, A.; Garcia, H. Eur. J. Inorg. Chem. 2004, 1143.
E
dx.doi.org/10.1021/cm502986d | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
(9) Baleizao, C.; Gigante, B.; Sabater, M. J.; Garcia, H.; Corma, A.
Appl. Catal., A 2002, 228, 279.
(30) (a) Fu, K.; Henbest, K.; Zhang, Y.; Valentin, S.; Sun, Y.-P. J.
Photochem. Photobiol., A 2002, 150, 143. (b) Gao, Y.; Song, Y.; Li, Y.;
Wang, Y.; Liu, H.; Zhu, D. Appl. Phys. B: Laser Opt. 2003, 76, 761.
(31) (a) Song, L.-C.; Liu, X.-F.; Xie, Z.-J.; Luo, F.-X.; Song, H.-B.
Inorg. Chem. 2011, 50, 11162. (b) Xu, W.; Feng, L.; Wu, Y.; Wang, T.;
Wu, J.; Xiang, J.; Li, B.; Jiang, L.; Shu, C.; Wang, C. Phys. Chem. Chem.
Phys. 2011, 13, 428. (c) D’Souza, F.; Gadde, S.; Islam, S.; Wijesinghe,
C.; Schumacher, A.; Zandler, M.; Araki, Y.; Ito, O. J. Phys. Chem. A
2007, 111, 8552. (d) Vail, S.; Krawczuk, P.; Guldi, D.; Palkar, A.;
(10) Bhattacharjee, S.; Anderson, J. A. Chem. Commun. 2004, 554.
(11) (a) Das, P.; Silva, A. R.; Carvalho, A. P.; Pires, J.; Freire, C.
Colloids Surf., A 2008, 329, 190. (b) Bai, R.; Fu, X.; Bao, H.; Ren, W.
Catal. Commun. 2008, 9, 1588. (c) Ren, W.; Fu, X.; Bao, H.; Bai, R.;
Ding, P.; Sui, B. Catal. Commun. 2009, 10, 788. (d) Gong, B.; Fu, X.;
Chen, J.; Li, Y.; Zou, X.; Tu, X.; Ding, P.; Ma, L. J. Catal. 2009, 262, 9.
(12) (a) Liu, G.; Wu, B.; Zhang, J.; Wang, X.; Shao, M.; Wang, J.
Inorg. Chem. 2009, 48, 2383. (b) Li, F.; Zhang, B.; Li, X.; Jiang, Y.;
Chen, L.; Li, Y.; Sun, L. Angew. Chem., Int. Ed. 2011, 50, 12276.
(c) Xing, L.; Xie, J.-H.; Chen, Y.-S.; Wang, L.-X.; Zhou, Q.-L. Adv.
Synth. Catal. 2008, 350, 1013. (d) Tran, P.; Le Goff, A.; Heidkamp, J.;
Jousselme, B.; Guillet, N.; Palacin, S.; Dau, H.; Fontecave, M.; Artero,
V. Angew. Chem., Int. Ed. 2011, 50, 1371. (e) Vriamont, C.; Devillers,
M.; Riant, O.; Hermans, S. Chem.Eur. J. 2013, 19, 12009.
(13) Chamberlain, T. W.; Popov, A. M.; Knizhnik, A. A.; Samoilov,
G. E.; Khlobystov, A. N. ACS Nano 2010, 4, 5203.
Echegoyen, L.; Tome,
11, 3375. (e) Wijesinghe, C.; El-Khouly, M.; Zandler, M.; Fukuzumi,
́
J.; Fazio, M.; Schuster, D. Chem.Eur. J. 2005,
S.; D’Souza, F. Chem.Eur. J. 2013, 19, 9629.
(14) Zulauf, A.; Mellah, M.; Hong, X.; Schulz, E. Dalton Trans. 2010,
39, 6911.
(15) Lebedeva, M. A.; Chamberlain, T. W.; Davies, E. S.; Mancel, D.;
Thomas, B. E.; Suyetin, M.; Bichoutskaia, E.; Schroder, M.;
̈
Khlobystov, A. N. Chem.Eur. J. 2013, 19, 11999.
(16) (a) Chamberlain, T. W.; Pfeiffer, R.; Peterlik, H.; Kuzmany, H.;
Zerbetto, F.; Melle-Franco, M.; Staddon, L.; Champness, N. R.; Briggs,
G. A. D.; Khlobystov, A. N. Small 2008, 4, 2262. (b) Chamberlain, T.
W.; Camenisch, A.; Champness, N. R.; Briggs, G. A. D.; Benjamin, S.
C.; Ardavan, A.; Khlobystov, A. N. J. Am. Chem. Soc. 2007, 129, 8609.
(c) Gimenez-Lopez, M. C.; Chuvilin, A.; Kaiser, U.; Khlobystov, A. N.
Chem. Commun. 2011, 47, 2116. (d) Chamberlain, T. W.; Champness,
N. R.; Schroder, M.; Khlobystov, A. N. Chem.Eur. J. 2011, 17, 668.
̈
(17) La Torre, A.; Fay, M. W.; Rance, G. A.; Gimen
́
ez-Lop
́
ez, M. C.;
Solomonsz, W. A.; Brown, P. D.; Khlobystov, A. N. Small 2012, 8,
1222.
(18) Solomonsz, W. A.; Rance, G. A.; Suyetin, M.; La Torre, A.;
Bichoutskaia, E.; Khlobystov, A. N. Chem.Eur. J. 2012, 18, 13180.
́ ́
(19) La Torre, A.; Gimenez-Lopez, M. C.; Fay, M. W.; Rance, G. A.;
Solomonsz, W. A.; Chamberlain, T. W.; Brown, P. D.; Khlobystov, A.
N. ACS Nano 2012, 6, 2000.
(20) Baleizao, C.; Gigante, B.; Garcia, H.; Corma, A. Tetrahedron
̃
2004, 60, 10461.
(21) Nooraeipour, M.; Moghadam, M.; Tangestaninejad, S.;
Mirkhani, V.; Mohammadpoor-Baltork, I.; Iravani, N. J. Coord.
Chem. 2012, 65, 226.
(22) Zakeri, M.; Moghadam, M.; Mohammadpoor-Baltork, I.;
Tangestaninejad, S.; Mirkhani, V.; Khosropour, A. R. J. Coord. Chem.
2012, 65, 1144.
(23) Liu, G.; Wu, B.; Zhang, J.; Wang, X.; Shao, M.; Wang, J. Inorg.
Chem. 2009, 48, 2383.
(24) Xing, L.; Xie, J.-H.; Chen, Y.-S.; Wang, L.-X.; Zhou, Q.-L. Adv.
Synth. Catal. 2008, 350, 1013.
(25) Su, H.; Li, Z.; Huo, Q.; Guan, J.; Kan, Q. RSC Adv. 2014, 4,
9990.
(26) (a) Minisci, F.; Fontana, F.; Araneo, S.; Recupero, F.; Banfi, S.;
Quici, S. J. Am. Chem. Soc. 1995, 117, 226. (b) Minisci, F.; Fontana, F.;
Araneo, S.; Rupero, F. J. Chem. Soc., Chem. Commun. 1994, 1823.
(c) Bravo, A.; Bjørsvik, H.-R.; Fontana, F.; Liguori, L.; Minisci, F. J.
Org. Chem. 1997, 62, 3849. (d) Yu, J.-Q.; Corey, E. Org. Lett. 2002, 4,
2727.
(27) Fraile, J. M.; Garcia, J. I.; Herrerias, C. I.; Mayoral, J. A.; Pires, E.
Chem. Soc. Rev. 2009, 38, 695.
(28) Meijer, M.; van Klink, G.; van Koten, G. Coord. Chem. Rev.
2002, 230, 141.
(29) (a) Maggini, M.; Guldi, D.; Mondini, S.; Scorrano, G.; Paolucci,
F.; Ceroni, P.; Roffia, S. Chem.Eur. J. 1998, 4, 1992. (b) Armaroli,
N.; Accorsi, G.; Felder, D.; Nierengarten, J.-F. Chem.Eur. J. 2002, 8,
2314. (c) Zhou, Z.; Sarova, G.; Zhang, S.; Ou, Z.; Tat, F.; Kadish, K.;
Echegoyen, L.; Guldi, D.; Schuster, D.; Wilson, S. Chem.Eur. J.
2006, 12, 4241.
F
dx.doi.org/10.1021/cm502986d | Chem. Mater. XXXX, XXX, XXX−XXX